Hi,
I have a bunch of wav files that I want to time align. I have coverted the wav files to raw format and tried using sphinx2-align.
But the problem is that I get the following error messages
time_align.c(460): rcsid == $Id: time_align.c,v 1.6 2001/02/13 19:51:38 lenzo Ex
p $
time_align.c(477): state_bp_table size 937K
time_align.c(481): phone_bp_table size 156K
time_align.c(485): word_bp_table size 15K
0 compound words found
time_align.c(1545): Increasing BPTable size from 15K by 15K
*** last state not reached at end of utterance ***
fbs_main.c(1949): No alignment for test1
The file format I'm using is raw 16 bit mono signed
Any suggestions would be very much helpful.
Thanks
Madan.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2002-03-26
Have a look at the FAQ web page. Have you tried time-align on the raw wav files ? or are you convinced that it takes raw PCM files.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2002-04-05
Alas no, havnt got all the way through the training sequence yet. Keeps bombing out @ CD tied stage. I think it has to do with the small amount of training data that i am using.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2002-06-13
Actually, I met simliar problem.
I successfully compiled and installed sphinx2 on my window 2000, and I wan to do some time-alignment for wav files.
Do I need to convert .wav file to raw format? how to do that? which tools?
Also, there is a file under ..\model\lm\turtle\goforward.16k which store audio data(I guess). how to convert my audio file(wav file) into this format?
or. I have to create my own phoetic models? how to do that? Thanks a lot.
-Realmman
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I have a bunch of wav files that I want to time align. I have coverted the wav files to raw format and tried using sphinx2-align.
But the problem is that I get the following error messages
time_align.c(460): rcsid == $Id: time_align.c,v 1.6 2001/02/13 19:51:38 lenzo Ex
p $
time_align.c(477): state_bp_table size 937K
time_align.c(481): phone_bp_table size 156K
time_align.c(485): word_bp_table size 15K
0 compound words found
time_align.c(1545): Increasing BPTable size from 15K by 15K
*** last state not reached at end of utterance ***
fbs_main.c(1949): No alignment for test1
The file format I'm using is raw 16 bit mono signed
Any suggestions would be very much helpful.
Thanks
Madan.
Have a look at the FAQ web page. Have you tried time-align on the raw wav files ? or are you convinced that it takes raw PCM files.
Hi!
Have you generated your own fonetic models?
Alas no, havnt got all the way through the training sequence yet. Keeps bombing out @ CD tied stage. I think it has to do with the small amount of training data that i am using.
Actually, I met simliar problem.
I successfully compiled and installed sphinx2 on my window 2000, and I wan to do some time-alignment for wav files.
Do I need to convert .wav file to raw format? how to do that? which tools?
Also, there is a file under ..\model\lm\turtle\goforward.16k which store audio data(I guess). how to convert my audio file(wav file) into this format?
or. I have to create my own phoetic models? how to do that? Thanks a lot.
-Realmman